The Impact of Model Complexity Reduction on System Dynamics: A Modelica Case Study of a Residential Heating System

Heat Pumps (HPs) powered by low-emission electricity from Renewable Energy Sources (RES) could contribute significantly to the decarbonization of the heating sector, moreover, integrating Domestic Hot Water (DHW) and Space Heating (SH) storages further enhances the demand-side flexibility for heating system operation. In this context, when simulating large-scale system models of e.g. residential heating systems, reducing complexity is necessary for numerical reasons. However, the accuracy of the simulation results must not be compromised. In the present paper, we investigate the effects of model complexity reduction on the system dynamics of residential heating systems utilizing the Modelica language. First, based on a real-world facility in Switzerland, we develop a most detailed model that consists of three main components: a HP, a SH storage and a DHW storage. We identify the potential of five possible simplifications of the model. Besides the reduction of the main components, we include the simplification of circulation pumps and the neglect of the pipes in the system. Our findings show that the DHW storage has the most impact on complexity reduction, whereas the HP simplification affects the system dynamics and CPU time of the simulation the most. With the most reduced model, we achieve a 41.3% reduction in complexity based on the number of unknowns and equations of the Differential-Algebraic System of Equations (DAE), the average CPU time reduction of 94.9% and even an improvement in model accuracy of 4% compared to the model with the highest level of detail.